Conviction Through Enhanced Fingerprint Identification
(The following is a case study which appeared in the December 1992 FBI / Law Enforcement Bulletin)
I n March 1990, an unknown assailant sexually molested and fatally stabbed a young woman. At the crime scene, an investigator discovered few leads. The only evidence was a pillowcase, found adjacent to the victim's body, that exhibited several bloodstains. One stain showed some faint fingerprint ridge detail, barely visible even to the trained eye.
An investigator took the pillowcase to the department's forensic unit for bloodstain pattern analysis. Technicians photographed and studied the stains, slowly extracting information. They discovered two things. First, they confirmed that several stains were consistent with blood transfer from a knife blade, although no knife was found at the crime scene. Second, and more importantly, analysts determined that the fingerprint presented enough ridge detail to conduct a more extensive investigation.
Analysts then sent the evidence to another forensic study center where scientists treated the fingerprint with DFO, a relatively new chemical (similar to Ninhydrin) that becomes fluorescent when exposed to a light source. Once processed, the DFO provided an improved ridge detail photo. However, the ridge detail still remained blurred, displaying poor general continuity and visible fabric weave in the background. All traditional photographic techniques failed to erase the distortion. Analysts subsequently concluded that the latent was unidentifiable.
A short time later, investigators assigned to the case witnessed a demonstration of fingerprint image enhancement at a forensic conference. Faced with a dead--end murder investigation, they decided to try the technique on the unidentifiable pillowcase fingerprint from the crime scene.
Investigators took the best DFO photograph and shipped it to a facility with the capability to perform image enhancement. Throughout the enhancement process, the accuracy of the print was documented through photographic records of each stage. Within 4 hours, the enhancement yielded an identifiable print.
In the interim, the lead case investigator developed several likely suspects. The primary suspect (the victim's next door neighbor) surfaced early in the investigation. However, the prints on record from a previous arrest did not contain sufficient ridge detail for comparison.
The investigator then concentrated on the serology report, which noted that examiners recovered seminal fluid from the victim during the postmortem examination. This preliminary serological report proved the seminal fluid matched that of the prime suspect, placing him in less than 5% of the general population. Encouraged by this breakthrough, examiners initiated the lengthy process of DNA analysis.
Using the serology report as probable cause for arrest, the investigators arrested the suspect and obtained a set of inked prints. After weeks of evaluation, comparison, and verification, the examiners achieved a positive identification comparison of the bloody pillow print with the left thumb of the suspect. Less than a week later, investigators received the DNA results, which further incriminated the suspect by matching his DNA code with that found in the stain on the pillowcase. This, in effect, placed the suspect as only 1 in 30 million people in the population with this particular DNA code.
During the suppression hearing, defense attorneys launched an attack on what they believed to be the most potentially vulnerable piece of evidence, the scientific acceptance of fingerprint image processing. To counter, an analyst took the court step by step through the entire procedure using a full complement of image enhancement equipment. An expert in the field of image processing then offered supporting testimony to the court.
Ultimately, the court ruled the enhanced print admissible, stating that the process did not alter the actual pattern of the print; it only made it more visible. The evidence passed the test, resulting in the first documented case where image enhancement technology withstood the challenges of a Frye hearing.1
One last piece of evidence emerged during final trial preparation. Maintenance men working in the defendant's vacant apartment discovered a military survival knife hidden in a pipe chase. Serological examination revealed traces of human blood, but no typing was possible. However, the shape and size of the saw tooth blade matched several of the blood stains on the pillowcase. Police personnel prepared a large transparent overlay for courtroom display to illustrate how the knife and the stain conformed to a single image.
Faced with overwhelming physical evidence, such as the image enhanced fingerprint match, the DNA test results, the match between the body fluid found on the victim's body and that of the suspect, and the knife found in the suspect's apartment, defense attorneys entered four guilty pleas, one of which was for capital murder. On June 18, 1991, the court sentenced the accused to four life sentences for murder and related offenses.2
Five years ago, a suspect committing these types of crimes would most likely go free, due to a lack of substantial forensic evidence. However, through persistence and by applying such modern technologies as fingerprint image enhancement, today's police investigators can use evidence invisible to their predecessors.
1 Frye v. United States, 293F.1013,1014 (D.C.Cir. 1923).
2 Commonwealth of Virginia v. Knight, CR--90--1353--02--F.
Information for this case study was submitted by Norman Tiller, a latent print examiner, and Thomas Tiller, a crime scene investigator, both with the Henrico County Division of Police, Richmond, Virginia.
(Editor --� For additional reading relating to this case study, see Journal of Forensic Identification 42 (2), pp 79--83. For additional reading which describes and explains the image enhancement technique utilized (the Fast Fourier Transform or �FFT�) see JFI 43 (6), pp 573--584. This landmark case has provided a solid foundation for future applications of Forensic Digital Image Enhancement.)
This article was reprinted in �THE PRINT� 10(2), February 1994, pp 1-2
and has been obtained from the online library provided by the
Southern California Association of Fingerprint Officers